It is common to read sheets (documents) by a scanner and store document image data representing document images. In order to store document images in a form convenient for a user, various image processing techniques of processing document image data have been proposed. For example, JP-A-2010-187056 describes a technique of using one document image data obtained by reading one page of a document having plural images printed thereon to generate plural image data representing the plural images, respectively.